2. Parametric and non parametric tests
Component 01
Research Methods
16 November 2018
Parametric and non parametric tests
L.O: To identify and describe the difference between parametric and non-parametric tests.

3. How will you know that you have made progress this lesson?
Progress Criteria
GOING STRONG
I can use the Mann Whitney U formula to calculate significance
ACCELERATING
I can explain different non-parametric tests and when to use them.
REVVING UP
I can define a parametric test and what they measure.

4. Statistical tests are classified into two types Parametric and Non-parametric
These are just a few of the statistical tests that psychologists use. However, the difference between the types of tests is based on the assumptions about the population and the type of data collated for analysis.
REVVING UP
Please note that the specification does not require knowledge of any specific parametric tests, all that is required, is the criteria for using them.

5. Assumptions of parametric tests
Populations drawn from should be normally distributed.
Variances of populations should be approximately equal.
Should have at least interval or ratio data.
Should be no extreme scores.
Remember there is no requirement within the specification to know specific types of parametric tests, how they are used, or any calculations.
REVVING UP

6. Anything else is non-parametric.
There are two types of test data and consequently different types of analysis.
As the table below shows, parametric data has an underlying normal distribution which allows for more conclusions to be drawn as the shape can be mathematically described.
Anything else is non-parametric.
REVVING UP
Parametric
Non Parametric
Assumed distribution
Normal
Any
Assumed variance
Homogeneous
Typical data
Ratio or Interval
Ordinal or nominal
Usual central tendency
Mean
Median
Can draw more conclusions
Simplicity – less affected by outliner scores

7. Understanding the use of non-parametric tests
Non-parametric tests are used for a variety of reasons, including:
When assumptions of the parametric tests cannot be fulfilled
When distributions are non-normal
ACCELERATING

8. Types of non-parametric tests
There are a number of non-parametric tests that can be used. These are:
Mann-Whitney U Test
Wilcoxon Signed Ranks test
Chi-Square
Binomial Sign Test
Spearman’s Rho.
Type of Data –do the findings from the study use nominal, ordinal or interval data?
Experimental Design – have you used independent measures or repeated measures design?
Differences in conditions – are you exploring differences in performance, test scores, between two conditions in your experimental study?
Are you looking for a relationship (or correlation) between the co-variables?
ACCELERATING

9. Key terms You will come across when carrying out statistical tests.
Observed Value – The number produced after the various steps and calculations for a statistical test have been carried out.
Critical Value – A value taken from a statistical test table, which must be reached in order for results to be significant.

11. Checklist for using the Mann Whitney U Test:
Tests to analyse the difference of two conditions of an IV: Ordinal or Interval Data. Independent Measures Design: Mann Whitney U Test
Checklist for using the Mann Whitney U Test:
DV produces ordinal or interval type of data
Independent Measures design
Exploring a difference between each condition (levels of the IV).
GOING STRONG

12. Mann Whitney U
Example A educational psychologist feels that men are better at abstract reasoning than women. To test this hypothesis, he collected the following percentile scores for 6 women and 7 men. The data is given below:
ho: There will be no difference between the score of men and women in an abstract reasoning test.
h1: There will score significantly more on a abstract reasoning test than women.
Women 81 80 50 95 93 85
Men 70 86 60 92 82 69 94

13. Rank each score
Rank 1 50 2 60 3 69 4 70 5 80 6 81 7 82 8 85 9 86 10 92 11 93 12 94 13 95
Rank (put in order from lowest to highest number) all scores together; ignore which groups the ranks are associated to.
Women 6 5 1 13 11 8
R1= 44
Men 4 9 2 10 7 3 12
R2 = 47
GOING STRONG

14. Working out U U = 44 – 6 x 7 / 2 6 x 7 = 42 42 / 2 = 21 44 – 21 = 23
You must use the smallest total value for this.
GOING STRONG

15. In order to determine whether this research is significant, the observed value of U has to be equal or less than the critical value of U.
Using table of critical U values for Mann-Whitney U Test.
The critical value is 6, meaning that the results are not significant.

16. Mann Whitney U
Example A educational psychologist feels that men and women will spend different amounts of time using their mobile phone.
Each participant was asked to estimate in an hour how many minutes they spend on their mobile.
Women 12 24 5 45 13 30 14
Men 6 10 16 35 20